Revert "Rasa"

madhavmk · Dec 30, 2019 · 0437161 · 0437161
1 parent 1c5f6c8
commit 0437161
Show file tree

Hide file tree

Showing 86 changed files with 1,031 additions and 1,239 deletions.
diff --git a/Audio_Outputs/__input_Clyde Tomb.wav b/Audio_Outputs/__input_Clyde Tomb.wav
diff --git a/Audio_Outputs/__input_Paris.wav b/Audio_Outputs/__input_Paris.wav
diff --git a/Audio_Outputs/__input_Streator, .wav b/Audio_Outputs/__input_Streator, .wav
diff --git a/Audio_Outputs/__input_€709 billi.wav b/Audio_Outputs/__input_€709 billi.wav
diff --git a/Data/nlu.md b/Data/nlu.md
diff --git a/Data/stories.md b/Data/stories.md
diff --git a/DeepSpeech/Models/info.txt b/DeepSpeech/Models/info.txt
diff --git a/README.md b/README.md
@@ -2,33 +2,6 @@
 
 This is an end-to-end voicebot that aims to answer open domain questions, and is intended to be used as a benchmarking tool
 
-## RASA Update
-Rasa update has been added in this branch. The additional dependencies are:
- - rasa (1.4.1)
- - rasa-sdk (1.4.0)
-
-Extra dependencies on Linux Systems (Tested on Ubuntu 18.04.3 LTS)
- - conda install -c conda-forge pygobject
- - pip install sanic==19.9.0
-
-Ensure that these are added to your venv
-
-The trained rasa model has already been added. The model responds to regular greetings, and when asked a question it redirects to the action server where the QA is performed. There are no additional intents.
-
-It has been set up as a client - server model. Details are as follows 
- - The main application is rasa_voice_client. It reads from the test_audio folder and runs the deepspeech model to generate the question
- - The rasa server on port 5002 handles conversation with the client
- - The rasa action server on port 5055 handles the custom action (Question answering engine is located here)
- - The T2S engine on port 5004 takes the answer as text input from the client and runs the Tacotron T2S on the text, and stores the audio files in the Audio_Output folder
-
-The parts are run with the following commands
- - Rasa server : rasa run -m models --endpoints endpoints.yml --port 5002 --credentials credentials.yml
- - Action server : rasa run actions
- - T2S : python rasa_t2s.py
- - Client : python rasa_voice_client.py
-
- Ensure that the client is started only after the other 3 servers have started
-
 ## Design
 **![](https://lh5.googleusercontent.com/2oFMv1ybATD_cmMO0CwzB-RAk6Nz-VG1wwDioIGWahLR4bVG51TIHbhHIUGTSpaLcVQS41QZIPOfX00VbZGCPa5O98st_VRsNlJnC3qEehpnEJrYLyLUOdCy-wiD34IC26wCac4KnxY)**
 
@@ -51,7 +24,6 @@ The parts are run with the following commands
  - matplotlib (3.1.1)
  - unidecode (1.1.1)
  - numpy (1.17.0)
-
 
 We recommend using a virtual environment to run this to prevent any conflicts with things like numpy.
 
@@ -81,14 +53,14 @@ To add your own audio to the testing set, simply place the wav file in the test-
 ## Running on Windows 10
 Run VoiceBot-windows.py
 Initially designed for the Windows platform. As such, all features should work perfectly.
-Outputs can be accessed from the '/Audio_outputs' subfolder.
+Outputs can be accessed from '/Vocoder_WaveRNN/WaveRNN_outputs' OR '/Tacotron_TTS/Tacotron_outputs' subfolders
 
 ## Running on Ubuntu
 Rin the VoiceBot-linux.py file
 playsound library and sounddevice library is not compatible.
 So, audio cannot be recorded or played on or from the console.
 VoiceBot can work only from questions pre-recorded in 'test_audio' folder.
-Outputs can be accessed from the '/Audio_outputs' subfolder.
+Outputs can be accessed from '/Vocoder_WaveRNN/WaveRNN_outputs' OR '/Tacotron_TTS/Tacotron_outputs' subfolders
 
 ## References
 
@@ -98,3 +70,5 @@ Outputs can be accessed from the '/Audio_outputs' subfolder.
  - [Fatchord's WaveRNN](https://github.com/fatchord/WaveRNN)
  - [BERT model trained by Surbhi Bhardwaj](https://github.com/surbhardwaj/BERT-QnA-Squad_2.0_Finetuned_Model)
 
+## Demo video
+Link to demo video here: https://drive.google.com/file/d/16pFeDjqDOCkVXW0cc09l_mkuxqgQjo8s/view?usp=drive_web
diff --git a/Tacotron_TTS/Tacotron_outputs/__input_365.wav b/Tacotron_TTS/Tacotron_outputs/__input_365.wav
diff --git a/Tacotron_TTS/Tacotron_outputs/temp.txt b/Tacotron_TTS/Tacotron_outputs/temp.txt
@@ -0,0 +1 @@
+again just a temp file
diff --git a/Tacotron_TTS/__init__.py b/Tacotron_TTS/__init__.py
diff --git a/Tacotron_TTS/hparams.py b/Tacotron_TTS/hparams.py
diff --git a/Tacotron_TTS/models/__init__.py b/Tacotron_TTS/models/__init__.py
diff --git a/Tacotron_TTS/models/helpers.py b/Tacotron_TTS/models/helpers.py
diff --git a/Tacotron_TTS/models/modules.py b/Tacotron_TTS/models/modules.py
diff --git a/Tacotron_TTS/models/rnn_wrappers.py b/Tacotron_TTS/models/rnn_wrappers.py
diff --git a/Tacotron_TTS/models/tacotron.py b/Tacotron_TTS/models/tacotron.py
diff --git a/Tacotron_TTS/models/temp.txt b/Tacotron_TTS/models/temp.txt
diff --git a/Tacotron_TTS/synthesizer.py b/Tacotron_TTS/synthesizer.py
diff --git a/Tacotron_TTS/tacotron_model_data/info.txt b/Tacotron_TTS/tacotron_model_data/info.txt
diff --git a/Tacotron_TTS/text/__init__.py b/Tacotron_TTS/text/__init__.py
diff --git a/Tacotron_TTS/text/cleaners.py b/Tacotron_TTS/text/cleaners.py
diff --git a/Tacotron_TTS/text/cmudict.py b/Tacotron_TTS/text/cmudict.py
diff --git a/Tacotron_TTS/text/numbers.py b/Tacotron_TTS/text/numbers.py
diff --git a/Tacotron_TTS/text/symbols.py b/Tacotron_TTS/text/symbols.py
diff --git a/Tacotron_TTS/text/temp.txt b/Tacotron_TTS/text/temp.txt
diff --git a/Tacotron_TTS/util/__init__.py b/Tacotron_TTS/util/__init__.py
diff --git a/Tacotron_TTS/util/audio.py b/Tacotron_TTS/util/audio.py
diff --git a/Tacotron_TTS/util/infolog.py b/Tacotron_TTS/util/infolog.py
diff --git a/Tacotron_TTS/util/plot.py b/Tacotron_TTS/util/plot.py
diff --git a/Vocoder_WaveRNN/WaveRNN_outputs/aud.txt b/Vocoder_WaveRNN/WaveRNN_outputs/aud.txt
@@ -0,0 +1 @@
+Generated audio files from wavernn + tacotron go here
diff --git a/Vocoder_WaveRNN/__init__.py b/Vocoder_WaveRNN/__init__.py
diff --git a/Vocoder_WaveRNN/info.txt b/Vocoder_WaveRNN/info.txt
diff --git a/Vocoder_WaveRNN/vocoder_hparams.py b/Vocoder_WaveRNN/vocoder_hparams.py
diff --git a/Vocoder_WaveRNN/vocoder_models/__init__.py b/Vocoder_WaveRNN/vocoder_models/__init__.py
diff --git a/Vocoder_WaveRNN/vocoder_models/deepmind_version.py b/Vocoder_WaveRNN/vocoder_models/deepmind_version.py
diff --git a/Vocoder_WaveRNN/vocoder_models/fatchord_version.py b/Vocoder_WaveRNN/vocoder_models/fatchord_version.py
diff --git a/Vocoder_WaveRNN/vocoder_models/tacotron.py b/Vocoder_WaveRNN/vocoder_models/tacotron.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/__init__.py b/Vocoder_WaveRNN/vocoder_utils/__init__.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/dataset.py b/Vocoder_WaveRNN/vocoder_utils/dataset.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/display.py b/Vocoder_WaveRNN/vocoder_utils/display.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/distribution.py b/Vocoder_WaveRNN/vocoder_utils/distribution.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/dsp.py b/Vocoder_WaveRNN/vocoder_utils/dsp.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/files.py b/Vocoder_WaveRNN/vocoder_utils/files.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/paths.py b/Vocoder_WaveRNN/vocoder_utils/paths.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/text/LICENSE b/Vocoder_WaveRNN/vocoder_utils/text/LICENSE
diff --git a/Vocoder_WaveRNN/vocoder_utils/text/__init__.py b/Vocoder_WaveRNN/vocoder_utils/text/__init__.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/text/cleaners.py b/Vocoder_WaveRNN/vocoder_utils/text/cleaners.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/text/cmudict.py b/Vocoder_WaveRNN/vocoder_utils/text/cmudict.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/text/numbers.py b/Vocoder_WaveRNN/vocoder_utils/text/numbers.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/text/recipes.py b/Vocoder_WaveRNN/vocoder_utils/text/recipes.py
diff --git a/Vocoder_WaveRNN/vocoder_utils/text/symbols.py b/Vocoder_WaveRNN/vocoder_utils/text/symbols.py
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		Generated audio files from wavernn + tacotron go here